On Tuesday, Google unveiled Nano Banana 2 Lite, the latest iteration of its proprietary AI for generating images and videos. The company asserts that this model is notably quicker and less costly compared to its predecessor.
With significantly reduced latency, the model can generate images within a mere four seconds, making it ideal for users who require rapid image production for various projects. Priced at just $0.034 per 1,000 images, it is a cost-effective solution for those looking to create and refine content in large quantities.
This release comes after last summer’s debut of the original Nano Banana, which utilized Gemini 3.1 Flash technology, and the February launch of Nano Banana 2 that added enhanced capabilities for producing more lifelike images. Google also provides Nano Banana Pro, touted as a more robust (and pricier) option for specialized applications.
While Nano Banana 2 is described as a “generalist workhorse,” the Lite version is specifically tailored for high-output environments where speed is critical, according to Google.
Despite some public dissatisfaction over the quality of AI-generated images, investments in AI-driven imagery and video creation remain strong. Google promotes its AI solutions as handy tools for crafting advertisements.
However, the relationship between the entertainment industry and AI developers is becoming increasingly complex, raising concerns among some creatives and audiences. Particularly notable is Google’s recent $75 million collaboration with indie studio A24, a partnership that has faced notable criticism from fans.
Available now through Google AI Studio, the Gemini API, and Google’s Gemini Enterprise Agent Platform, Nano Banana 2 Lite supersedes the older Nano Banana model, which the company now classifies as a “legacy model.”
In addition, Google announced a broader rollout of Gemini Omni Flash, first previewed during the Google I/O event earlier this year. This system is priced at $0.10 per second of video output. Furthermore, Google introduced Omni Product Studio, designed to convert static images generated by the Omni system into “cinematic e-commerce videos.”
“Creating with generative media often involves creative iterations,” the company stated in a recent blog post. “With these two models, developers are equipped to create comprehensive multimedia experiences that seamlessly link fast image generation with video creation and editing.”
